PDA

View Full Version : What is the Most Accurate Way



Brian R.
21st September 2004, 10:43
to measure the distance you travel while driving?

Odometers are notoriously inaccurate.

Mileage markers? How were the placing of mileage markers on highways determined? By an old fart in a pick-up truck with his odometer? Every time it hit a 0 in the tenths, he dropped a stick out the window?

This is important!

dZeus
21st September 2004, 10:46
a palmtop loaded with gps software + receiver?

Wombat
21st September 2004, 10:51
Odometers are notoriously inaccurate?

Brian R.
21st September 2004, 10:52
Will that give you actual distance traveled along a curved path? or only the line of sight distance between two points along the route?

Brian R.
21st September 2004, 10:54
Originally posted by Wombat
Odometers are notoriously inaccurate?
That's the word

dZeus
21st September 2004, 11:07
Originally posted by Brian R.
Will that give you actual distance traveled along a curved path? or only the line of sight distance between two points along the route?

afaik, there's software to do that (along the travelled path). TomTom maybe? (I don't have first hand experience with it)

Helevitia
21st September 2004, 11:10
My GPS works really well :)

My odometer is off because I have bigger tires on my truck. My GPS gives me much better accuracy.

How accurate do you want it to be?

Fluff
21st September 2004, 11:15
My Tom Tom reads 30mph and the speedo reads 34mph.

Odometers are probably with tolerances of +/- 10%

Also Trip computers are probably quite inaccurate. when they determine mpg.

I wonder how accurate gas station meters are? I wouldn't be suprised if they were designed to be calibrated to the low end of the tolerance for 1 gallon of gas.

KRSESQ
21st September 2004, 11:30
I don't know who's in charge of calibrating gas pumps across the pond, but around here gas pumps are inspected annually by the Public Service Commission, whose requirements for accuracy are quite stringent.

Public Service Commission members in North Dakota are elected rather than appointed, so charges of cronyism and corruption are minimized and public accountability is maximized.

As long as the tire size for your vehicle is correct, the speedometer and odometer should be accurate to within 1 or 2 percent. If you have odd size tires, all bets are off.

At least, that's what my owner's manual says. :rolleyes:

If you can find a garage with a dynamometer, I'm sure they'd be willing to test it for you, for a price. :D

Kevin

Brian R.
21st September 2004, 11:33
Originally posted by Helevitia
How accurate do you want it to be?

I was thinking in the range where you can accurately determine your gasoline mileage to say +/- 1.0 mpg or about 4%. If the error in the gas pumps is too high, I guess that's not possible.

As far as the accuracy of odometers go, a guy I was talking to said his workplace was either 27 or 31 miles from his home depending on which of his cars he drove, both Toyotas. That's a discrepancy of somewhere around 15%.

Greebe
21st September 2004, 11:38
Gasoline meters are calibrated at set volumes. That being said there was a scandle not so many years ago where these were intentionally calibrated to read more at anything but the standards ie if you were to measure out 1, 5, 10 gallons it read properly, but if you measured out anything other than these amounts it would always give you less gas. If memory serves me correct there was several chain of gasoline stations busted on this in California along with the companies who supplied the altered metering controls.

KRSESQ
21st September 2004, 11:43
Also the temp of the gasoline has to be taken into account. Gasoline expands quite a bit when warm, so even if the pump accurately meters out one gallon, you'll still be getting less gas than you would if the gas were cold.

Most gasoline storage tanks are underground, though, so the temp is reasonably constant throughout the year.

Kevin

Fluff
21st September 2004, 11:47
Just looked at the UK regulations. It allows for a +/- 60ml error in 12Litres of fuel which is pretty stringent.

I'm not sure if the wheels for my car (peugeot)are the original shipping ones or whether the previous owner bought some new peugeot alloys which were a slightly different size.


On fuel temp there is this study..

http://www.nel.uk/upload/pdfs/warm-fuel.pdf

GNEP
21st September 2004, 12:08
Reminds me of the endless discussions we have with our operating companies in different climates on how much volume they sell... correcting to STP conditions is not a concept most accountants who actually record this stuff understand :)

But hey, when I work in kbbl/day at best what do I care about the odd gallon here or there :p

Jon P. Inghram
21st September 2004, 12:32
The odometer on my ratty old '87 Cavalier was dead-on accurate (verfied via GPS,) at least until I bought new tires that had a different profile. Now it's off a few percent, at 65 MPH it reads ~63 for example.

Paddy
21st September 2004, 12:57
I don't know about accuracy, but my odometer is consistent to less than 1%.

spadnos
21st September 2004, 13:59
Barring mechanical failures, an odometer should be perfectly consistent. (This is a little less true with electronic odometers.) A mechanical odometer is geared from the output shaft of the transmission (or one of the axles), and so turns at a fixed ratio to the wheel speed.

The accuracy is mostly determined by tire size. If you use big tires, you'll be going faster than the speedometer tells you, and the odometer will be accumulating fewer miles than you actually travel.

- Steve

degrub
21st September 2004, 19:43
Tyre size and to some extent pressure, make the most difference. An electronic odometer is counting gear teeth rotating by or the equivalent and will be as accurate as any mechanical odometer. As far as the volumne of petrol you meter into your car, the meter is typically calibrated to the average temperature of delivered petrol so the error produced is very small, less than .005 parts or .5% and more likely half that.

The other variable is the fill pipe "tank full" flow cut off. Can easily vary by .5 gallons or more across pumps. And the cuttoff will vary with pumping rate. If you want to be more accurate, use the same pump each time and measure the mileage over several fillups.

DukeP
21st September 2004, 22:27
The most effective way is to calculate it.
Your RPM gauge is precise, its actually counting the revolutions.
Know the gearratio and the diameters of your tire - no matter how much air you inflate it with, if you stay within the norm psi, its not going to affect your calculations (much)
Its all down to Pi. :)

Thats how its done when driving cars that go FAST, at least.

Used it to clock in an old BMW. Made 320kmh - which even on the German Freeway is kind of scary.

~~DukeP~~

Brian R.
21st September 2004, 23:42
How about a fifth wheel?

Marshmallowman
22nd September 2004, 02:14
"Just" find an accurately marked distance of road,

Then calculate the correction factor using the odo reading and the known distance.

Here in oz we have very accurately marked distance markers on some country roads (1km) use for arial speed check's.

Fat Tone
22nd September 2004, 04:09
Originally posted by GNEP
...But hey, when I work in kbbl/day at best what do I care about the odd gallon here or there :p


That's precisely when you should care! :) A small cc error on one litre or barrel can soon add up on that scale!

In the UK it's ultimately administered by NWML (National Weights and Measures Ltd. They maintain the national standards for legal metrology. I once turned down a job with them...I remember a few petrol pumps were in the labs for type-testing when I had my interview.

As to the accuracy of speedos, there was a report out ~1 year ago that said that on average they over-read a little at 30mph and underread a little at 70mph. If you look at the mechanism, it's not surprising that big deviations are reported: One I took apart because the needle was oscillating wildly consisted of a spinning magnet causing a 'stator' to deflect, which was being heavily damped by a coiled spring.

GNEP
22nd September 2004, 04:23
Originally posted by Fat Tone
That's precisely when you should care! :) A small cc error on one litre or barrel can soon add up on that scale!
Nah - I work on the principle that globally all the errors up and down pretty much average out.

Anyway, the plan numbers I'm looking at now are all imaginary anyway...

and it would appear that the stockmarket is not anticipating hearing what they want to hear at our strategy presentation this afternoon... oh well.

Novdid
22nd September 2004, 04:53
Originally posted by Fat Tone
That's precisely when you should care! :) A small cc error on one litre or barrel can soon add up on that scale!

In the UK it's ultimately administered by NWML (National Weights and Measures Ltd. They maintain the national standards for legal metrology. I once turned down a job with them...I remember a few petrol pumps were in the labs for type-testing when I had my interview.

As to the accuracy of speedos, there was a report out ~1 year ago that said that on average they over-read a little at 30mph and underread a little at 70mph. If you look at the mechanism, it's not surprising that big deviations are reported: One I took apart because the needle was oscillating wildly consisted of a spinning magnet causing a 'stator' to deflect, which was being heavily damped by a coiled spring.

Let me guess, Ford Sierra???:p

Fat Tone
22nd September 2004, 05:04
Actually it was a Vauxhall! :)

Paddy
22nd September 2004, 06:20
You suprised me with your knowledge there T! :)